Large amount of information from various sources is readily available over the Internet and new sources of information are being introduced daily. The current trend has lead to a large combination of information bases that could be physically miles apart but contain information that need to be combined for accurate response to a request. In order for the user to be able to extract useful information that first, match as closely as possible with their requirements; and secondly is provided fast or within a minimal time delay, various strategies from the design of memories and access methods to database architecture design and inference rules and guidelines have been developed. In many occasions the information extraction rules need to be modified. For example, if information bases containing contradictory rules are combined, the contradiction issue needs to be addressed before user queries return wrong answers. Furthermore, modification of reasoning systems or deduction engines may be needed in order to achieve more accurate answers to queries imposed by information base users. Currently in the existing systems sets of deduction engines and inference rules are statically introduced in the system and any modifications must be performed by authorized system administrators. However, this process could slow down the information extraction process.